X-Git-Url: https://git.karo-electronics.de/?a=blobdiff_plain;f=drivers%2Fmd%2Fmd.c;h=3276edde757644244248af7e6e2435c8cfd7e5db;hb=b73df2d3d629aefa187a0a3574fd81455e026bc8;hp=df1230af02cd5cb3db9f8bdc1771841d707b4011;hpb=0cd17fec983b6bca505eecee1af33138687220b6;p=mv-sheeva.git

diff --git a/drivers/md/md.c b/drivers/md/md.c
index df1230af02c..3276edde757 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -347,21 +347,20 @@ static struct mdk_personality *find_pers(int level, char *clevel)
 	return NULL;
 }
 
+/* return the offset of the super block in 512byte sectors */
 static inline sector_t calc_dev_sboffset(struct block_device *bdev)
 {
-	sector_t size = bdev->bd_inode->i_size >> BLOCK_SIZE_BITS;
-	return MD_NEW_SIZE_BLOCKS(size);
+	sector_t num_sectors = bdev->bd_inode->i_size / 512;
+	return MD_NEW_SIZE_SECTORS(num_sectors);
 }
 
-static sector_t calc_dev_size(mdk_rdev_t *rdev, unsigned chunk_size)
+static sector_t calc_num_sectors(mdk_rdev_t *rdev, unsigned chunk_size)
 {
-	sector_t size;
-
-	size = rdev->sb_offset;
+	sector_t num_sectors = rdev->sb_offset * 2;
 
 	if (chunk_size)
-		size &= ~((sector_t)chunk_size/1024 - 1);
-	return size;
+		num_sectors &= ~((sector_t)chunk_size/512 - 1);
+	return num_sectors;
 }
 
 static int alloc_disk_sb(mdk_rdev_t * rdev)
@@ -372,7 +371,7 @@ static int alloc_disk_sb(mdk_rdev_t * rdev)
 	rdev->sb_page = alloc_page(GFP_KERNEL);
 	if (!rdev->sb_page) {
 		printk(KERN_ALERT "md: out of memory.\n");
-		return -EINVAL;
+		return -ENOMEM;
 	}
 
 	return 0;
@@ -543,17 +542,12 @@ fail:
 
 static int uuid_equal(mdp_super_t *sb1, mdp_super_t *sb2)
 {
-	if (	(sb1->set_uuid0 == sb2->set_uuid0) &&
-		(sb1->set_uuid1 == sb2->set_uuid1) &&
-		(sb1->set_uuid2 == sb2->set_uuid2) &&
-		(sb1->set_uuid3 == sb2->set_uuid3))
-
-		return 1;
-
-	return 0;
+	return 	sb1->set_uuid0 == sb2->set_uuid0 &&
+		sb1->set_uuid1 == sb2->set_uuid1 &&
+		sb1->set_uuid2 == sb2->set_uuid2 &&
+		sb1->set_uuid3 == sb2->set_uuid3;
 }
 
-
 static int sb_equal(mdp_super_t *sb1, mdp_super_t *sb2)
 {
 	int ret;
@@ -564,7 +558,7 @@ static int sb_equal(mdp_super_t *sb1, mdp_super_t *sb2)
 
 	if (!tmp1 || !tmp2) {
 		ret = 0;
-		printk(KERN_INFO "md.c: sb1 is not equal to sb2!\n");
+		printk(KERN_INFO "md.c sb_equal(): failed to allocate memory!\n");
 		goto abort;
 	}
 
@@ -577,11 +571,7 @@ static int sb_equal(mdp_super_t *sb1, mdp_super_t *sb2)
 	tmp1->nr_disks = 0;
 	tmp2->nr_disks = 0;
 
-	if (memcmp(tmp1, tmp2, MD_SB_GENERIC_CONSTANT_WORDS * 4))
-		ret = 0;
-	else
-		ret = 1;
-
+	ret = (memcmp(tmp1, tmp2, MD_SB_GENERIC_CONSTANT_WORDS * 4) == 0);
 abort:
 	kfree(tmp1);
 	kfree(tmp2);
@@ -684,7 +674,7 @@ static int super_90_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version
 	 *
 	 * It also happens to be a multiple of 4Kb.
 	 */
-	sb_offset = calc_dev_sboffset(rdev->bdev);
+	sb_offset = calc_dev_sboffset(rdev->bdev) / 2;
 	rdev->sb_offset = sb_offset;
 
 	ret = read_disk_sb(rdev, MD_SB_BYTES);
@@ -762,7 +752,7 @@ static int super_90_load(mdk_rdev_t *rdev, mdk_rdev_t *refdev, int minor_version
 		else 
 			ret = 0;
 	}
-	rdev->size = calc_dev_size(rdev, sb->chunk_size);
+	rdev->size = calc_num_sectors(rdev, sb->chunk_size) / 2;
 
 	if (rdev->size < sb->size && sb->level > 1)
 		/* "this cannot possibly happen" ... */
@@ -1017,7 +1007,7 @@ super_90_rdev_size_change(mdk_rdev_t *rdev, unsigned long long size)
 	size *= 2; /* convert to sectors */
 	if (rdev->mddev->bitmap_offset)
 		return 0; /* can't move bitmap */
-	rdev->sb_offset = calc_dev_sboffset(rdev->bdev);
+	rdev->sb_offset = calc_dev_sboffset(rdev->bdev) / 2;
 	if (!size || size > rdev->sb_offset*2)
 		size = rdev->sb_offset*2;
 	md_super_write(rdev->mddev, rdev, rdev->sb_offset << 1, rdev->sb_size,
@@ -1849,7 +1839,7 @@ repeat:
 
 }
 
-/* words written to sysfs files may, or my not, be \n terminated.
+/* words written to sysfs files may, or may not, be \n terminated.
  * We want to accept with case. For this we use cmd_match.
  */
 static int cmd_match(const char *cmd, const char *str)
@@ -2624,7 +2614,7 @@ __ATTR(resync_start, S_IRUGO|S_IWUSR, resync_start_show, resync_start_store);
  *     When written, doesn't tear down array, but just stops it
  * suspended (not supported yet)
  *     All IO requests will block. The array can be reconfigured.
- *     Writing this, if accepted, will block until array is quiessent
+ *     Writing this, if accepted, will block until array is quiescent
  * readonly
  *     no resync can happen.  no superblocks get written.
  *     write requests fail
@@ -2697,7 +2687,7 @@ array_state_show(mddev_t *mddev, char *page)
 	return sprintf(page, "%s\n", array_states[st]);
 }
 
-static int do_md_stop(mddev_t * mddev, int ro);
+static int do_md_stop(mddev_t * mddev, int ro, int is_open);
 static int do_md_run(mddev_t * mddev);
 static int restart_array(mddev_t *mddev);
 
@@ -2713,14 +2703,14 @@ array_state_store(mddev_t *mddev, const char *buf, size_t len)
 		/* stopping an active array */
 		if (atomic_read(&mddev->active) > 1)
 			return -EBUSY;
-		err = do_md_stop(mddev, 0);
+		err = do_md_stop(mddev, 0, 0);
 		break;
 	case inactive:
 		/* stopping an active array */
 		if (mddev->pers) {
 			if (atomic_read(&mddev->active) > 1)
 				return -EBUSY;
-			err = do_md_stop(mddev, 2);
+			err = do_md_stop(mddev, 2, 0);
 		} else
 			err = 0; /* already inactive */
 		break;
@@ -2728,7 +2718,7 @@ array_state_store(mddev_t *mddev, const char *buf, size_t len)
 		break; /* not supported yet */
 	case readonly:
 		if (mddev->pers)
-			err = do_md_stop(mddev, 1);
+			err = do_md_stop(mddev, 1, 0);
 		else {
 			mddev->ro = 1;
 			set_disk_ro(mddev->gendisk, 1);
@@ -2738,7 +2728,7 @@ array_state_store(mddev_t *mddev, const char *buf, size_t len)
 	case read_auto:
 		if (mddev->pers) {
 			if (mddev->ro != 1)
-				err = do_md_stop(mddev, 1);
+				err = do_md_stop(mddev, 1, 0);
 			else
 				err = restart_array(mddev);
 			if (err == 0) {
@@ -2899,7 +2889,7 @@ size_show(mddev_t *mddev, char *page)
 	return sprintf(page, "%llu\n", (unsigned long long)mddev->size);
 }
 
-static int update_size(mddev_t *mddev, unsigned long size);
+static int update_size(mddev_t *mddev, sector_t num_sectors);
 
 static ssize_t
 size_store(mddev_t *mddev, const char *buf, size_t len)
@@ -2916,7 +2906,7 @@ size_store(mddev_t *mddev, const char *buf, size_t len)
 		return -EINVAL;
 
 	if (mddev->pers) {
-		err = update_size(mddev, size);
+		err = update_size(mddev, size * 2);
 		md_update_sb(mddev, 1);
 	} else {
 		if (mddev->size == 0 ||
@@ -3168,11 +3158,11 @@ static ssize_t
 sync_speed_show(mddev_t *mddev, char *page)
 {
 	unsigned long resync, dt, db;
-	resync = (mddev->curr_mark_cnt - atomic_read(&mddev->recovery_active));
-	dt = ((jiffies - mddev->resync_mark) / HZ);
+	resync = mddev->curr_mark_cnt - atomic_read(&mddev->recovery_active);
+	dt = (jiffies - mddev->resync_mark) / HZ;
 	if (!dt) dt++;
-	db = resync - (mddev->resync_mark_cnt);
-	return sprintf(page, "%ld\n", db/dt/2); /* K/sec */
+	db = resync - mddev->resync_mark_cnt;
+	return sprintf(page, "%lu\n", db/dt/2); /* K/sec */
 }
 
 static struct md_sysfs_entry md_sync_speed = __ATTR_RO(sync_speed);
@@ -3675,7 +3665,9 @@ static int do_md_run(mddev_t * mddev)
 		mddev->ro = 2; /* read-only, but switch on first write */
 
 	err = mddev->pers->run(mddev);
-	if (!err && mddev->pers->sync_request) {
+	if (err)
+		printk(KERN_ERR "md: pers->run() failed ...\n");
+	else if (mddev->pers->sync_request) {
 		err = bitmap_create(mddev);
 		if (err) {
 			printk(KERN_ERR "%s: failed to create bitmap (%d)\n",
@@ -3684,7 +3676,6 @@ static int do_md_run(mddev_t * mddev)
 		}
 	}
 	if (err) {
-		printk(KERN_ERR "md: pers->run() failed ...\n");
 		module_put(mddev->pers->owner);
 		mddev->pers = NULL;
 		bitmap_destroy(mddev);
@@ -3774,40 +3765,25 @@ static int do_md_run(mddev_t * mddev)
 static int restart_array(mddev_t *mddev)
 {
 	struct gendisk *disk = mddev->gendisk;
-	int err;
 
-	/*
-	 * Complain if it has no devices
-	 */
-	err = -ENXIO;
+	/* Complain if it has no devices */
 	if (list_empty(&mddev->disks))
-		goto out;
-
-	if (mddev->pers) {
-		err = -EBUSY;
-		if (!mddev->ro)
-			goto out;
-
-		mddev->safemode = 0;
-		mddev->ro = 0;
-		set_disk_ro(disk, 0);
-
-		printk(KERN_INFO "md: %s switched to read-write mode.\n",
-			mdname(mddev));
-		/*
-		 * Kick recovery or resync if necessary
-		 */
-		set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
-		md_wakeup_thread(mddev->thread);
-		md_wakeup_thread(mddev->sync_thread);
-		err = 0;
-		sysfs_notify(&mddev->kobj, NULL, "array_state");
-
-	} else
-		err = -EINVAL;
-
-out:
-	return err;
+		return -ENXIO;
+	if (!mddev->pers)
+		return -EINVAL;
+	if (!mddev->ro)
+		return -EBUSY;
+	mddev->safemode = 0;
+	mddev->ro = 0;
+	set_disk_ro(disk, 0);
+	printk(KERN_INFO "md: %s switched to read-write mode.\n",
+		mdname(mddev));
+	/* Kick recovery or resync if necessary */
+	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
+	md_wakeup_thread(mddev->thread);
+	md_wakeup_thread(mddev->sync_thread);
+	sysfs_notify(&mddev->kobj, NULL, "array_state");
+	return 0;
 }
 
 /* similar to deny_write_access, but accounts for our holding a reference
@@ -3841,16 +3817,17 @@ static void restore_bitmap_write_access(struct file *file)
  *   1 - switch to readonly
  *   2 - stop but do not disassemble array
  */
-static int do_md_stop(mddev_t * mddev, int mode)
+static int do_md_stop(mddev_t * mddev, int mode, int is_open)
 {
 	int err = 0;
 	struct gendisk *disk = mddev->gendisk;
 
+	if (atomic_read(&mddev->active) > 1 + is_open) {
+		printk("md: %s still in use.\n",mdname(mddev));
+		return -EBUSY;
+	}
+
 	if (mddev->pers) {
-		if (atomic_read(&mddev->active)>2) {
-			printk("md: %s still in use.\n",mdname(mddev));
-			return -EBUSY;
-		}
 
 		if (mddev->sync_thread) {
 			set_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
@@ -3999,7 +3976,7 @@ static void autorun_array(mddev_t *mddev)
 	err = do_md_run (mddev);
 	if (err) {
 		printk(KERN_WARNING "md: do_md_run() returned %d\n", err);
-		do_md_stop (mddev, 0);
+		do_md_stop (mddev, 0, 0);
 	}
 }
 
@@ -4172,9 +4149,11 @@ static int get_bitmap_file(mddev_t * mddev, void __user * arg)
 	char *ptr, *buf = NULL;
 	int err = -ENOMEM;
 
-	md_allow_write(mddev);
+	if (md_allow_write(mddev))
+		file = kmalloc(sizeof(*file), GFP_NOIO);
+	else
+		file = kmalloc(sizeof(*file), GFP_KERNEL);
 
-	file = kmalloc(sizeof(*file), GFP_KERNEL);
 	if (!file)
 		goto out;
 
@@ -4207,15 +4186,12 @@ out:
 static int get_disk_info(mddev_t * mddev, void __user * arg)
 {
 	mdu_disk_info_t info;
-	unsigned int nr;
 	mdk_rdev_t *rdev;
 
 	if (copy_from_user(&info, arg, sizeof(info)))
 		return -EFAULT;
 
-	nr = info.number;
-
-	rdev = find_rdev_nr(mddev, nr);
+	rdev = find_rdev_nr(mddev, info.number);
 	if (rdev) {
 		info.major = MAJOR(rdev->bdev->bd_dev);
 		info.minor = MINOR(rdev->bdev->bd_dev);
@@ -4381,8 +4357,8 @@ static int add_new_disk(mddev_t * mddev, mdu_disk_info_t *info)
 			printk(KERN_INFO "md: nonpersistent superblock ...\n");
 			rdev->sb_offset = rdev->bdev->bd_inode->i_size >> BLOCK_SIZE_BITS;
 		} else 
-			rdev->sb_offset = calc_dev_sboffset(rdev->bdev);
-		rdev->size = calc_dev_size(rdev, mddev->chunk_size);
+			rdev->sb_offset = calc_dev_sboffset(rdev->bdev) / 2;
+		rdev->size = calc_num_sectors(rdev, mddev->chunk_size) / 2;
 
 		err = bind_rdev_to_array(rdev, mddev);
 		if (err) {
@@ -4421,7 +4397,6 @@ static int hot_add_disk(mddev_t * mddev, dev_t dev)
 {
 	char b[BDEVNAME_SIZE];
 	int err;
-	unsigned int size;
 	mdk_rdev_t *rdev;
 
 	if (!mddev->pers)
@@ -4449,13 +4424,12 @@ static int hot_add_disk(mddev_t * mddev, dev_t dev)
 	}
 
 	if (mddev->persistent)
-		rdev->sb_offset = calc_dev_sboffset(rdev->bdev);
+		rdev->sb_offset = calc_dev_sboffset(rdev->bdev) / 2;
 	else
 		rdev->sb_offset =
 			rdev->bdev->bd_inode->i_size >> BLOCK_SIZE_BITS;
 
-	size = calc_dev_size(rdev, mddev->chunk_size);
-	rdev->size = size;
+	rdev->size = calc_num_sectors(rdev, mddev->chunk_size) / 2;
 
 	if (test_bit(Faulty, &rdev->flags)) {
 		printk(KERN_WARNING 
@@ -4640,24 +4614,24 @@ static int set_array_info(mddev_t * mddev, mdu_array_info_t *info)
 	return 0;
 }
 
-static int update_size(mddev_t *mddev, unsigned long size)
+static int update_size(mddev_t *mddev, sector_t num_sectors)
 {
 	mdk_rdev_t * rdev;
 	int rv;
 	struct list_head *tmp;
-	int fit = (size == 0);
+	int fit = (num_sectors == 0);
 
 	if (mddev->pers->resize == NULL)
 		return -EINVAL;
-	/* The "size" is the amount of each device that is used.
-	 * This can only make sense for arrays with redundancy.
-	 * linear and raid0 always use whatever space is available
-	 * We can only consider changing the size if no resync
-	 * or reconstruction is happening, and if the new size
-	 * is acceptable. It must fit before the sb_offset or,
-	 * if that is <data_offset, it must fit before the
-	 * size of each device.
-	 * If size is zero, we find the largest size that fits.
+	/* The "num_sectors" is the number of sectors of each device that
+	 * is used.  This can only make sense for arrays with redundancy.
+	 * linear and raid0 always use whatever space is available. We can only
+	 * consider changing this number if no resync or reconstruction is
+	 * happening, and if the new size is acceptable. It must fit before the
+	 * sb_offset or, if that is <data_offset, it must fit before the size
+	 * of each device.  If num_sectors is zero, we find the largest size
+	 * that fits.
+
 	 */
 	if (mddev->sync_thread)
 		return -EBUSY;
@@ -4665,12 +4639,12 @@ static int update_size(mddev_t *mddev, unsigned long size)
 		sector_t avail;
 		avail = rdev->size * 2;
 
-		if (fit && (size == 0 || size > avail/2))
-			size = avail/2;
-		if (avail < ((sector_t)size << 1))
+		if (fit && (num_sectors == 0 || num_sectors > avail))
+			num_sectors = avail;
+		if (avail < num_sectors)
 			return -ENOSPC;
 	}
-	rv = mddev->pers->resize(mddev, (sector_t)size *2);
+	rv = mddev->pers->resize(mddev, num_sectors);
 	if (!rv) {
 		struct block_device *bdev;
 
@@ -4752,7 +4726,7 @@ static int update_array_info(mddev_t *mddev, mdu_array_info_t *info)
 			return mddev->pers->reconfig(mddev, info->layout, -1);
 	}
 	if (info->size >= 0 && mddev->size != info->size)
-		rv = update_size(mddev, info->size);
+		rv = update_size(mddev, (sector_t)info->size * 2);
 
 	if (mddev->raid_disks    != info->raid_disks)
 		rv = update_raid_disks(mddev, info->raid_disks);
@@ -4805,6 +4779,12 @@ static int set_disk_faulty(mddev_t *mddev, dev_t dev)
 	return 0;
 }
 
+/*
+ * We have a problem here : there is no easy way to give a CHS
+ * virtual geometry. We currently pretend that we have a 2 heads
+ * 4 sectors (with a BIG number of cylinders...). This drives
+ * dosfs just mad... ;-)
+ */
 static int md_getgeo(struct block_device *bdev, struct hd_geometry *geo)
 {
 	mddev_t *mddev = bdev->bd_disk->private_data;
@@ -4949,19 +4929,13 @@ static int md_ioctl(struct inode *inode, struct file *file,
 			goto done_unlock;
 
 		case STOP_ARRAY:
-			err = do_md_stop (mddev, 0);
+			err = do_md_stop (mddev, 0, 1);
 			goto done_unlock;
 
 		case STOP_ARRAY_RO:
-			err = do_md_stop (mddev, 1);
+			err = do_md_stop (mddev, 1, 1);
 			goto done_unlock;
 
-	/*
-	 * We have a problem here : there is no easy way to give a CHS
-	 * virtual geometry. We currently pretend that we have a 2 heads
-	 * 4 sectors (with a BIG number of cylinders...). This drives
-	 * dosfs just mad... ;-)
-	 */
 	}
 
 	/*
@@ -4971,14 +4945,12 @@ static int md_ioctl(struct inode *inode, struct file *file,
 	 * here and hit the 'default' below, so only disallow
 	 * 'md' ioctls, and switch to rw mode if started auto-readonly.
 	 */
-	if (_IOC_TYPE(cmd) == MD_MAJOR &&
-	    mddev->ro && mddev->pers) {
+	if (_IOC_TYPE(cmd) == MD_MAJOR && mddev->ro && mddev->pers) {
 		if (mddev->ro == 2) {
 			mddev->ro = 0;
 			sysfs_notify(&mddev->kobj, NULL, "array_state");
 			set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
 			md_wakeup_thread(mddev->thread);
-
 		} else {
 			err = -EROFS;
 			goto abort_unlock;
@@ -5667,15 +5639,18 @@ void md_write_end(mddev_t *mddev)
  * may proceed without blocking.  It is important to call this before
  * attempting a GFP_KERNEL allocation while holding the mddev lock.
  * Must be called with mddev_lock held.
+ *
+ * In the ->external case MD_CHANGE_CLEAN can not be cleared until mddev->lock
+ * is dropped, so return -EAGAIN after notifying userspace.
  */
-void md_allow_write(mddev_t *mddev)
+int md_allow_write(mddev_t *mddev)
 {
 	if (!mddev->pers)
-		return;
+		return 0;
 	if (mddev->ro)
-		return;
+		return 0;
 	if (!mddev->pers->sync_request)
-		return;
+		return 0;
 
 	spin_lock_irq(&mddev->write_lock);
 	if (mddev->in_sync) {
@@ -5686,14 +5661,14 @@ void md_allow_write(mddev_t *mddev)
 			mddev->safemode = 1;
 		spin_unlock_irq(&mddev->write_lock);
 		md_update_sb(mddev, 0);
-
 		sysfs_notify(&mddev->kobj, NULL, "array_state");
-		/* wait for the dirty state to be recorded in the metadata */
-		wait_event(mddev->sb_wait,
-			   !test_bit(MD_CHANGE_CLEAN, &mddev->flags) &&
-			   !test_bit(MD_CHANGE_PENDING, &mddev->flags));
 	} else
 		spin_unlock_irq(&mddev->write_lock);
+
+	if (test_bit(MD_CHANGE_CLEAN, &mddev->flags))
+		return -EAGAIN;
+	else
+		return 0;
 }
 EXPORT_SYMBOL_GPL(md_allow_write);
 
@@ -6249,7 +6224,7 @@ static int md_notify_reboot(struct notifier_block *this,
 
 		for_each_mddev(mddev, tmp)
 			if (mddev_trylock(mddev)) {
-				do_md_stop (mddev, 1);
+				do_md_stop (mddev, 1, 0);
 				mddev_unlock(mddev);
 			}
 		/*